Objectives: To identify determinants of incident malnutrition in community-dwelling older adults. Design: Meta-analysis of 6 community-based longitudinal datasets with follow-up of 1 to 3 years. Setting: Datasets from MaNuEL (MalNutrition in the Elderly) partners were included: 3 studies from Germany and 1 each from Ireland, the Netherlands, and New Zealand. Participants: community-dwelling adults aged 65 and older (N=4,844). Measurement: The same definition of incident malnutrition was used for all cohorts (body mass index < 20.0 kg/m2 at follow-up or weight loss ≥10 % between baseline and follow-up). Twenty-one potential baseline determinants from 7 domains (demographic, nutritional, lifestyle, social, psychological, physical functioning, medical) and 2 follow-up variables (hospitalization, falls) were harmonized for all studies. Binary logistic regression analyses were performed to assess the association between each variable, adjusted for specific confounders, and incident malnutrition. Combined odds ratios (ORs) with 95% confidence intervals (CIs) were calculated using random-effects meta-analyses. Results: Studies included between 209 and 1,841 participants without malnutrition at baseline; mean age ranged from 71.7 to 84.6. Incidence of malnutrition varied from 5.1% and 17.2%. Meta-analyses identified 6 variables as independent determinants of incident malnutrition; with increasing age, the risk of developing malnutrition increased continuously. Unmarried, separated, or divorced participants were more likely to develop malnutrition than married participants, whereas no association was found for widowed participants. Participants with difficulty walking (OR=1.41, 95% CI=1.06–1.89) or difficulty climbing stairs (OR=1.45, 95% CI=1.14–1.85) and those who were hospitalized before baseline (OR=1.49, 95% CI=1.25–1.76) and during follow-up (OR=2.02, 95% CI=1.41–2.88) had higher odds of incident malnutrition. Conclusion: In this harmonized meta-analysis based on prospective data of older, community-dwelling adults, age, marital status, limitations with walking and climbing stairs, and hospitalization were identified as determinants of incident malnutrition. J Am Geriatr Soc 66:2335–2343, 2018.
- longitudinal study